Upload
others
View
12
Download
0
Embed Size (px)
Citation preview
Fiscal Uncertainty and How to Deal with It
Alan AuerbachUniversity of California, Berkeley
January 29, 2019
Basic Conundrum
• For many countries, long-term projections for the paths of deficits look much worse than those for the very short term, in part because of demographic factors– An unsustainable path without major changes from current policy;
i.e. a large fiscal gap
• But the size of this gap is subject to considerable uncertainty• How should policy deal with the prospect?• Comments draw on Auerbach (2014) and other papers
Uncertainty and the Decision Horizon
• Even in the simpler case of short-term policy decisions, there is considerable uncertainty– Illustration: US 10-year forecast and ex post federal budget deficits
Uncertainty and the Decision Horizon
• Even in the simpler case of short-term policy decisions, there is considerable uncertainty– Illustration: US 10-year forecast and ex post federal budget deficits
• Moreover, error bands increase rapidly with forecast horizon, due to compounding errors for many sources of uncertainty– Illustration: US federal budget deficits, holding policy given, as
projected by the US Congressional Budget Office at the outset of the global financial crisis and ex post
Uncertainty and the Decision Horizon
• For longer-term forecasts, considerable uncertainty• Typically not expressed in terms of confidence bounds,
because of difficulty even of defining them• Also, some sources of long-term uncertainty not necessarily
subject to the same stochastic properties over time
What to Do?
• Issue of interest here is not how to deal with fiscal imbalances, but how our responses should change because of this uncertainty– As with OASDI, imbalances may be likely, but we don’t know how
large
What to Do?
• Issue of interest here is not how to deal with fiscal imbalances, but how our responses should change because of this uncertainty– As with OASDI, imbalances may be likely, but we don’t know how
large– Also, policy itself, both through reactions to economic conditions and
independent shocks, is a main source of uncertainty
Source: Auerbach et al. (2018)
Common Responses
• Stein’s Law– “If something cannot go on forever, it will stop”– Implication: if, under some circumstances, projected deficits are
unsustainable, then they won’t continue
Common Responses
• Stein’s Law– “If something cannot go on forever, it will stop”– Implication: if, under some circumstances, projected deficits are
unsustainable, then they won’t continue
• Changes will occur, but when and how?• And how will such a “passive” trajectory compare to a
desirable, planned transition?
Common Responses
• Stein’s Law• Projections beyond [ ] years are so uncertain that we should
ignore them, or at least discount them more in our policy responses
Common Responses
• Stein’s Law• Projections beyond [ ] years are so uncertain that we should
ignore them, or at least discount them more in our policy responses
• But what does that imply?– Assume problems that are projected to be big and getting worse as of
year t vanish in year t+1?
Taking Precaution
• If the future is very uncertain, a case for saving more for the future, whether by an individual or by a government
• Things more complicated when analyzing what government should do, but the main conclusion still holds, and may even be reinforced
Government vs. Individual Saving
• Government saving decisions involve aggregation over individual cohorts– If people will be better off in the future, they can absorb greater fiscal
burdens
Government vs. Individual Saving
• Government saving decisions involve aggregation over individual cohorts– If people will be better off in the future, they can absorb greater fiscal
burdens
• But, this is an argument about how to deal with projected fiscal imbalance, not with the uncertainty about it– E.g., perhaps save less now to account for future well-being, but still
want a precautionary response
Government vs. Individual Saving
• Also, governments must use distortionary taxation to raise resources– Economic costs of very high future marginal tax rates push toward
more active responses now, because1) Larger dispersion of future outcomes2) Worse average future outcome
• A similar motivation comes from an incentive to avoid a costly fiscal crisis if adversity strikes
Further Issues
Risk vs. Uncertainty
• Knightian uncertainty– E.g., it’s not that we are very uncertain about the future; we really
have no idea
• This type of uncertainty may be something we wish to avoid, but not something we should ignore– We may possibly want to be even more prudent in response
Waiting to Learn More
• Suppose we will have a better idea about the trajectory in the future
• Should we wait to act?
Waiting to Learn More
• Suppose we will have a better idea about the trajectory in the future
• Should we wait to act?• It depends on the evolution of uncertainty
– If new shocks appear at a steady rate, there is no gain to waiting, and a loss due to the restriction of our options
– Even if important types of uncertainty will be resolved, this should influence the types of responses, rather than our decision to act
Costs of Political Action
• Perhaps it is difficult to gather consensus to modify policy; we can’t be changing policy continually whenever information changes
Costs of Political Action
• Perhaps it is difficult to gather consensus to modify policy; we can’t be changing policy continually whenever information changes– Implies that we may wish to wait to act, with a bias toward less fiscal
tightening (Auerbach and Hassett, 2007)
Costs of Political Action
• Perhaps it is difficult to gather consensus to modify policy; we can’t be changing policy continually whenever information changes– Implies that we may wish to wait to act, with a bias toward less fiscal
tightening (Auerbach and Hassett, 2007)– But when we do act, we should act more forcefully
• We can also put in place some automatic responses if we can be fairly confident about what those responses should be as events unfold
Political Pressure on Forecasts
• With greater uncertainty, projections are more susceptible to political influence (e.g., Auerbach 1999)
Political Pressure on Forecasts
• With greater uncertainty, projections are more susceptible to political influence (e.g., Auerbach 1999)
• This may well be true– Some evidence of overly optimistic projections in some environments
• But this provides an argument for institutional protections and transparency, not for ignoring information
Summary
• Uncertainty means our policy choices will always turn out to be “wrong”
• But ignoring uncertainty doesn’t make it go away; a more active response can lessen its negative consequences